Quantitative bone marrow magnetic resonance imaging through apparent diffusion coefficient and fat fraction in multiple myeloma patients

• Published in: Radiologia medica Vol. 126; no. 3; pp. 445 - 452
• Main Authors: Berardo, Sara, Sukhovei, Lidiia, Andorno, Silvano, Carriero, Alessandro, Stecco, Alessandro
• Format: Journal Article
• Language: English
• Published: Milan Springer Milan 01.03.2021; Springer Nature B.V
• Subjects: Adipose Tissue - diagnostic imaging; Adult; Aged; Aged, 80 and over; Analysis of Variance; Body Water - diagnostic imaging; Bone marrow; Bone Marrow - diagnostic imaging; Case-Control Studies; Diagnostic Radiology; Diffusion coefficient; Diffusion Magnetic Resonance Imaging - methods; Disease Progression; Female; Humans; Imaging; Infiltration; Interventional Radiology; Magnetic resonance imaging; Male; Mathematical analysis; Medicine; Medicine & Public Health; Middle Aged; Multiple myeloma; Multiple Myeloma - diagnostic imaging; Musculoskeletal Radiology; Neuroradiology; Pelvis; Quantitative analysis; Radiology; Retrospective Studies; Sensitivity; Sensitivity and Specificity; Spine - diagnostic imaging; Ultrasound; ADC; Fat fraction; Whole body MRI; Multiple myeloma; DWI; Chemical shift imaging
• ISSN: 0033-8362; 1826-6983; 1826-6983
• DOI: 10.1007/s11547-020-01258-z

Objective Quantitative bone marrow (BM) MR sequences, as DWI and CSI, were used to evaluate BM water–fat composition. The aim of the study was to assess the potential usefulness of fat fraction (FF) and ADC, calculated by CSI or DWI, in diagnosing and classifying myeloma (MM) patients according to their different BM infiltration patterns. Methods The study group included 43 MM patients (19F; 24M; mean age 64 years), 15 asymptomatic, 15 symptomatic with diffuse BM infiltration and 13 symptomatic with focal lesions (FLs). The control group was made up of 15 healthy subjects (7F; 8M; mean age 64 years). MRI examinations consisted of sagittal T1w TSE on the spinal column, axial DWI ( b 50–400–800 mm 2 /s) and coronal T2 Dixon, on the whole body. Mean ADC and FF were calculated placing 1 ROI on 6 vertebras and 2 ROIs on either the pelvis or FL. Results ANOVA with Bonferroni’s correction showed a significant difference in ADC values among the different groups of MM patients ( P  < 0.05), while FF was only significantly different between patients with diffuse infiltration and patients with FL ( P  = 0.002). ADC allowed distinguishing MM patients from normal BM patients with diffuse BM infiltration (cutoff value: 0.491 × 10 −3  mm 2 /s; sensitivity 73%, specificity 80%). FF helped better discriminate healthy controls from normal BM patients (cutoff = 0.33, sensitivity 73%, specificity 92%) and patients with diffuse BM infiltration from those with FL (cutoff = 0.16, sensitivity 82%, specificity 92%). Conclusion ADC and FF are potentially useful parameter for the quantitative evaluation of BM infiltration in MM patients.